It’s Sunday morning, and you’re deciding what to make for brunch. You could just make something up or look through a cookbook, but instead you decide to let your computer do the work. And so it does, using its knowledge of what flavors appeal to you on a molecular level to spit out a recipe for a healthy meal that has just that perfect amount of sweet and salty for your finicky taste buds.
That’s not science fiction; it’s a scenario that could become reality in just five years. Every year, IBM puts out its 5 in 5, a list of five innovations that could change our lives in the next five years. This year, the company focused on the ways that cognitive computing–where computers learn through experience like humans–will change the world. Watson, the IBM supercomputer that handily beat two Jeopardy champions, uses cognitive computing. The 2012 predictions look at how future cognitive computers will learn through the human senses: touch, sight, hearing, taste, smell.
And that’s where the recipe-developing computer comes in. “At the end of the day, taste is about chemistry. The nose is an extraordinarily sensitive chemical sensor, says Bernie Meyerson, VP of innovation at IBM. A computing system currently being developed by IBM can break down food to the molecular level, using today’s knowledge about flavor and smell combinations that people enjoy to create entirely new–sometimes weird–recipes. The example that IBM gives is pairing roasted chestnuts with cooked beetroot, caviar, or dry-cured ham.
The computer could theoretically go even further than that, customizing foods for individual preferences. Meyerson explains: “You can tell the computer over course of a year, give it a thumbs up or thumbs down with each thing you eat. You collect all that data, and the computer can continuously run an analysis and look for common threads from the chemical makeup of what’s pleasurable [to you].” Then, Meyerson says, the computer could optimize food within the bounds of what’s healthy. It’s possible that companies could use the technology on a broad scale as well to figure out how to produce tasty, healthy products that people will enjoy.
It may seem fanciful, but Meyerson believes that the technology predicted in the latest 5 in 5–digital taste buds, computers that correlate baby sounds with physiological information, systems with a sense of smell that can diagnose diseases, and more–will be “extremely inexpensive” in five years. “We have the modern technology for sensors required to carry this off,” he says.